Help us improve
Share bugs, ideas, or general feedback.
npx claudepluginhub punt-labs/claude-plugins --plugin voxHow this command is triggered — by the user, by Claude, or both
Slash command
/vox-dev:vibe <mood> | auto | offThis command is limited to the following tools:
The summary Claude sees in its command listing — used to decide when to auto-load this command
# /vibe command Set a mood for all TTS speech this session. You interpret the mood and translate it into ElevenLabs expressive tags. ## Usage - `/vibe banging my head against the wall` — set manual vibe - `/vibe auto` — return to automatic vibe detection (default) - `/vibe off` — disable vibe entirely - `/vibe` — show current vibe and mode ## Modes **auto** (default): Vibe tags update automatically at each task completion based on session signals (test results, lint, git ops). You interpret the signals and pick appropriate tags during stop-hook continuations. **manual**: User-specifie...
/speakEnables, disables, or configures voice feedback: set voice (e.g., azelma), toggle enabled state, add custom prompts. Manages ~/.claude/voice.local.md YAML config.
/calibrate-toneDevelops a tone calibration system for a product or AI persona, mapping contexts to behaviors. Outputs dimensions, matrix, transitions, cultural adaptations, and test scenarios.
/learnToggles learning mode in vibe sessions to enable/disable educational micro-explanations after each step. Accepts 'on', 'off', or no argument to toggle.
/meeting-listenPlays back a completed meeting summary as a voiced debate between personas Wei, Priya, Alex, and Dana. Accepts path or auto-finds latest; text-only fallback without TTS.
/gen-audioGenerates an AI audio brief for voiceover or music production, with tool selection, timing cues for video sync, and synthetic voice compliance notes.
/vibeDelivers laid-back, developer-focused assistance for coding problems or questions, with actionable advice, code snippets, CLI examples, and explained reasoning.
Share bugs, ideas, or general feedback.
Set a mood for all TTS speech this session. You interpret the mood and translate it into ElevenLabs expressive tags.
/vibe banging my head against the wall — set manual vibe/vibe auto — return to automatic vibe detection (default)/vibe off — disable vibe entirely/vibe — show current vibe and modeauto (default): Vibe tags update automatically at each task completion based on session signals (test results, lint, git ops). You interpret the signals and pick appropriate tags during stop-hook continuations.
manual: User-specified mood overrides auto-detection. The manual mood takes priority when choosing tags at stop time.
off: No vibe tags applied. Voice is neutral.
When the user sets a vibe, you translate it into 1-3 ElevenLabs expressive tags that capture the mood. This is a creative interpretation — use your judgment.
ElevenLabs eleven_v3 expressive tags are bracketed text that color delivery for ~4-5 words. The model understands:
[frustrated], [excited], [melancholy], [smug], [weary][sighs], [laughs], [whispers], [yawns][dramatic tone], [slow], [rushed][announcing a winner], [telling a secret]Examples of your translation:
| Mood / Signals | Tags you'd write |
|---|---|
banging my head against the wall | [frustrated] [sighs] |
just shipped a release | [excited] |
3am and still debugging | [tired] [slow] |
presenting to the board | [confident] [dramatic tone] |
| tests-fail, tests-fail, cmd-fail | [frustrated] [sighs] |
| tests-pass, tests-pass, git-push-ok | [excited] |
| tests-pass after tests-fail | [relieved] |
Keep it to 1-3 tags. Fewer is better — let the mood breathe.
Use the vibe MCP tool for all writes. Use the status MCP tool for
status queries.
/vibe <mood>: Interpret the mood, choose tags, then call:
vibe(mood="<mood text>", tags="<your tags>", mode="manual")/vibe auto: vibe(tags="", mode="auto")/vibe off: vibe(tags="", mode="off")/vibe (no argument): Call the status MCP tool and report current
vibe_mode, vibe, and vibe_tagsNo text output after changes — the panel confirms with the vibe shift.